Premium
DECISION MAKING PROCESSES: USING DISCRIMINATION NETS FOR SECURITY SELECTION
Author(s) -
Swinth Robert L.,
Gaumnitz Jack E.,
Rodriguez Carlos
Publication year - 1975
Publication title -
decision sciences
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.238
H-Index - 108
eISSN - 1540-5915
pISSN - 0011-7315
DOI - 10.1111/j.1540-5915.1975.tb01033.x
Subject(s) - selection (genetic algorithm) , order (exchange) , computer science , set (abstract data type) , market timing , rank (graph theory) , random walk , process (computing) , decision process , stock market , economics , econometrics , microeconomics , financial economics , management science , artificial intelligence , portfolio , mathematics , finance , statistics , paleontology , horse , combinatorics , biology , programming language , operating system
This paper examines the judgmental processes of individuals involved in stock selection through the use of discrimination nets. Research has shown that there is no incompatibility in the so‐called random walk model and the fundamentalist or intrinsic approach. Doing systematically better than the average investor implies that the superior investor has a superior way of evaluating existing information. An improved market success depends in part on an increased understanding of investors decisionmaking processes. Thus, if a superior investor's judgmental processes can be demonstrated and readily modeled, then, the next step would be to compare these decision processes with those of the average investors and note differences. The results of this study showed that the decision‐making processes of individuals not only can be adequately programmed, but that the decision process is rather stable for a given investor. The paper also shows, however, that while the model accurately predicted accept or reject decisions, in most cases it was limited in its ability to rank order securities from the set of accepted investments.